Smart Video Number Plate Character Recognition Using Hybrid Optimization-Based YoloV3

The number plate recognition (NPR) system plays a vital role in solving multi-level problems of security. NPR utilizes hybrid model techniques comprising image processing and optical character recognition (OCR). Accuracy measures how well the ANPR system correctly identifies license plate characters. Precision-recall balance is crucial in ensuring that the system does not miss any important details (recall) while minimizing false positives (precision). By combining precision and recall in a single metric, the F1-score provides a complete evaluation of the system’s performance. Efficient management and monitoring in areas like toll roads and parking lots rely heavily on the reliability of ANPR systems. They enable streamlined operations, enhance security, and facilitate effective enforcement of regulations. NPR enables efficient management and monitoring in various applications like parking facilities for both public and private sectors, restricted areas such as gated communities, secure facilities, and sensitive government installations. Primary phase of NPR system is to capture image of number plate by using camera or imaging devices. By utilizing image processing techniques quality of image is enhanced and once the image is enhanced next step is locating the number plate within the image. The next step is to apply OCR algorithm to image to accurately identify and recognize number plate character. ANPR system accuracy depends on two important parameters, namely, quality of captured image and efficiency of OCR algorithm. Survey of research papers shows that the accuracy of recognizing Indian number using NPR system is 75–85% because of font size, regional language, and environmental conditions. Use of Optimized YOLOv3 for detecting number plate on high-speed vehicle is logical choice. YOLOv3 is a popular algorithm for its speed and accuracy. By Improved Convolutional Neural Network (ICNN), character recognition of number plate is done after identifying the number plate. Overall, better performance has shown by Convolutional Neural Network (CNNs) in the identifying process. Convolutional Neural Networks (CNNs) have demonstrated strong performance in image recognition tasks. The role of NPR systems is vital in enhancing efficiency, safety, and compliance in transportation law enforcement and parking management system. Also, it is helpful for automatic toll collection. Overall, the NPR system has the ability to improve efficiency, enhance security, ensure compliance, enhance customer experience, and, by integrating with other technologies, make it a vital assets for transportation-related operations.